<junari>
It is still necessary to make a frequency separation for different bins, but it doesn't affect the work
<junari>
I have a second device on t527 in addition to the old one on h728, now it will be easier to do it and I will do it as soon as I have more free time for that
junari_ has joined #linux-sunxi
junari has quit [Remote host closed the connection]
<smaeul>
apritzel: just tested H6 and it also allows access to the first 0x40 eFuse bytes from non-secure state
<smaeul>
apritzel: would it make sense to replace needs_smc_workaround_if_zero_word_at_addr with checking the CPSR or something? that seems like it would be more reliable than trying to find a memory location with the desired properties
<smaeul>
btw the SMC trick as implemented doesn't work on secure H6 either
cnxsoft has joined #linux-sunxi
Daanct12 has joined #linux-sunxi
hexdump0815 has joined #linux-sunxi
hexdump02 has quit [Ping timeout: 480 seconds]
hexdump02 has joined #linux-sunxi
bauen1 has quit [Ping timeout: 480 seconds]
hexdump01 has quit [Ping timeout: 480 seconds]
jakllsch has quit [Read error: Connection reset by peer]
bauen1 has joined #linux-sunxi
jakllsch has joined #linux-sunxi
radxanaoki has joined #linux-sunxi
dsimic is now known as Guest13129
dsimic has joined #linux-sunxi
Guest13129 has quit [Ping timeout: 480 seconds]
IlikeTech has quit [Ping timeout: 480 seconds]
Daanct12 has quit [Read error: Connection reset by peer]
Daanct12 has joined #linux-sunxi
IlikeTech has joined #linux-sunxi
ungeskriptet_ has joined #linux-sunxi
ungeskriptet has quit [Ping timeout: 480 seconds]
Daanct12 has quit [Ping timeout: 480 seconds]
bauen1 has quit [Ping timeout: 480 seconds]
JohnDoe_71Rus has joined #linux-sunxi
Daanct12 has joined #linux-sunxi
IlikeTech has quit [Ping timeout: 480 seconds]
IlikeTech has joined #linux-sunxi
Robot_ has joined #linux-sunxi
apritzel_ has joined #linux-sunxi
bauen1 has joined #linux-sunxi
Schimsalabim has quit [Ping timeout: 480 seconds]
Schimsalabim has joined #linux-sunxi
gestalt has joined #linux-sunxi
gestalt has quit [Remote host closed the connection]
gestalt has joined #linux-sunxi
gestalt has quit [Remote host closed the connection]
gestalt has joined #linux-sunxi
gestalt_ has joined #linux-sunxi
junari_ has quit [Remote host closed the connection]
junari_ has joined #linux-sunxi
<gestalt_>
Hey I wanted to ask where I can find the allwinner d1s sdk? I can't seem to find it anywhere
gestalt_ has quit [Remote host closed the connection]
gestalt has quit [Ping timeout: 480 seconds]
colinsane has joined #linux-sunxi
Schimsalabim has quit [Ping timeout: 480 seconds]
Schimsalabim has joined #linux-sunxi
Schimsalabim has quit [Read error: Connection reset by peer]
Schimsalabim has joined #linux-sunxi
gestalt has joined #linux-sunxi
Schimsalabim has quit [Ping timeout: 480 seconds]
Schimsalabim has joined #linux-sunxi
colinsane has quit [Ping timeout: 480 seconds]
warpme has joined #linux-sunxi
Schimsalabim has quit [Read error: Connection reset by peer]
Schimsalabim has joined #linux-sunxi
apritzel_ has quit [Ping timeout: 480 seconds]
warpme has quit []
evgeny_boger1 has joined #linux-sunxi
<apritzel>
smaeul: yes, we surely need to replace this detection mechanism, I guess this only works on the A64 ;-) It's just not easy regardless. Most of the time we are in SVC, so CPSR is of no help either.
<apritzel>
but I am hopeful to find some kind of hint, would probably be very SoC specific
<apritzel>
smaeul: and thanks for testing the H6!
<apritzel>
I saw it not working on my secure A133 tablet either, I guess now I have some more clues to follow...
<apritzel>
junari_: any chance we can see some of those patches on the list? Doesn't need to be perfect, but we need to start somewhere, and now (-rc1 being out) is a good time
evgeny_boger has quit [Ping timeout: 480 seconds]
<apritzel>
I guess we can iron out the CPU CCU driver pretty quickly, for instance, as that's pretty contained and independent from the rest
<apritzel>
similarly the thermal driver can be looked at in isolation first
junari_ is now known as junari
<junari>
Yes, I think I might do that this weekend
bauen1 has quit [Ping timeout: 480 seconds]
gestalt_ has joined #linux-sunxi
warpme has joined #linux-sunxi
<gestalt_>
Is there a way to download the allwinner f133-a (d1s) sdk (rtos) somewhere?
<Lightsword>
apritzel, jernej, btw the ram module on the h616 I have that's not working in mainline should be a model K4B2G1646F-BCNB
gestalt_ has quit [Remote host closed the connection]
Schimsalabim has quit [Ping timeout: 480 seconds]
Schimsalabim has joined #linux-sunxi
<apritzel>
Lightsword: ah, yes, that's by 16. Do you know if there is only one chip or two?
<apritzel>
scratch that, must be one, since that's already the 256MB
<Lightsword>
apritzel, yeah, should only be one
<apritzel>
so the driver seems to support half-width, just not sure that has ever been tested
warpme has quit []
junari has quit [Remote host closed the connection]
<apritzel>
hilarious how the Samsung page for that chip starts with "Developed in 2005, ..." ;-)
junari has joined #linux-sunxi
montjoie1 has quit [Ping timeout: 480 seconds]
montjoie2 has quit [Ping timeout: 480 seconds]
warpme has joined #linux-sunxi
bauen1 has joined #linux-sunxi
gestalt has quit [Read error: Connection reset by peer]
junari has quit [Remote host closed the connection]
Daanct12 has quit [Read error: Connection reset by peer]
montjoie has joined #linux-sunxi
<apritzel>
junari: many thanks, that would be great! Don't worry if there are issues, we can sort that out on the list
Daanct12 has joined #linux-sunxi
Daanct12 has quit [Quit: WeeChat 4.6.0]
JohnDoe_71Rus has joined #linux-sunxi
junari_ has joined #linux-sunxi
junari has quit [Remote host closed the connection]
cnxsoft has quit [Ping timeout: 480 seconds]
Schimsalabim has quit [Ping timeout: 480 seconds]
Schimsalabim has joined #linux-sunxi
warpme has quit []
junari_ has quit [Remote host closed the connection]
evgeny_boger has joined #linux-sunxi
evgeny_boger1 has quit [Read error: Connection reset by peer]
palmer has quit [Read error: Connection reset by peer]
narmstrong has quit [Read error: No route to host]
steev has quit [Write error: connection closed]
NishanthMenon has quit [Read error: No route to host]
arnd has quit [Read error: No route to host]
lvrp16_ has quit [Read error: No route to host]
palmer has joined #linux-sunxi
revy_ has quit [Read error: No route to host]
lvrp16_ has joined #linux-sunxi
arnd has joined #linux-sunxi
narmstrong has joined #linux-sunxi
NishanthMenon has joined #linux-sunxi
revy_ has joined #linux-sunxi
steev has joined #linux-sunxi
<Lightsword>
apritzel, so the driver currently using this "DRAM:testing 32-bit width, rank = 2" is a bug probably?
<apritzel>
why?
<Lightsword>
apritzel, you said the ram chip in the board is 16 bit not 32 right?
<apritzel>
those checks mean we have provisions to support single and dual rank, and full (32) and half width (16 bits). I just don't know if we have seen a device using only 16-bits, so this might be untested
<apritzel>
as the comment says, it detects those things at runtime, and starts with the widest possible
<smaeul>
Lightsword: out of curiosity, does the TOC0 still boot if you remove the `uint8_t __pad[2];` member from toc0_nvc_seq in mkimage? it should, but I don't have a device to test with
<smaeul>
it'll only load two bytes of the NVC, but that should be fine because the max value is 31 (the number of bits in the eFuse) anyway
<smaeul>
btw the code for parsing this extension has an unbounded memcpy using the length of the sequence, which is most likely exploitable :-D
<smaeul>
s/unbounded memcpy/& onto the stack/g
<Lightsword>
smaeul, yep, still passes
<smaeul>
Lightsword: great, thanks!
<smaeul>
I'm not sure how to plumb this into upstream mkimage, since I don't think older chips can handle the extra data there. I suppose it could look for a file in the key directory containing the counter value
<Lightsword>
smaeul, we can't just max out the sequence or something?
<smaeul>
Lightsword: IIRC older chips look for the signature sequence exactly 0x20 bytes after the start of the firmware digest
<Lightsword>
smaeul, so h616 is jumping over that field conditionally?
<smaeul>
yes
<Lightsword>
smaeul, maybe needs a cli param to set sequence?
<Lightsword>
and have that wired up for configuration in kconfig
<smaeul>
we don't really control what parameters mkimage has. I was thinking a file in the key directory: only insert the sequence if the file exists
apritzel has quit [Ping timeout: 480 seconds]
Schimsalabim has quit [Read error: Connection reset by peer]